Purpose Our aim was to develop and validate a machine learning (ML)-based approach for interpretation of I-123 FP-CIT SPECT scans to discriminate Parkinson's disease (PD) from non-PD and to determine its generalizability and clinical value in two centers. Methods We retrospectively included 210 consecutive patients who underwent I-123 FP-CIT SPECT imaging and had a clinically confirmed diagnosis. Linear support vector machine (SVM) was used to build a classification model to discriminate PD from non-PD based on I-123-FP-CIT striatal uptake ratios, age and gender of 90 patients. The model was validated on unseen data from the same center where the model was developed (n = 40) and consecutively on data from a different center (n = 80). Prediction performance was assessed and compared to the scan interpretation by expert physicians. ResultsTesting the derived SVM model on the unseen dataset (n = 40) from the same center resulted in an accuracy of 95.0%, sensitivity of 96.0% and specificity of 93.3%. This was identical to the classification accuracy of nuclear medicine physicians. The model was generalizable towards the other center as prediction performance did not differ thereby obtaining an accuracy of 82.5%, sensitivity of 88.5% and specificity of 71.4% (p = NS). This was comparable to that of nuclear medicine physicians (p = NS). Conclusion ML-based interpretation of I-123-FP-CIT scans results in accurate discrimination of PD from non-PD similar to visual assessment in both centers. The derived SVM model is therefore generalizable towards centers using comparable acquisition and image processing methods and implementation as diagnostic aid in clinical practice is encouraged.

Background. Clinical practice shows degrading image quality in heavier patients who undergo myocardial perfusion imaging (MPI) with Rubidium-82 (Rb-82) PET when using a fixed tracer activity. Our aim was to derive and validate a patient-specific activity protocol resulting in a constant image quality in PET MPI.Methods. We included 251 patients who underwent rest MPI with Rb-82 PET (Discovery 670, GE Healthcare). 132 patients were included retrospectively and were scanned using a fixed activity of 740 MBq. The total number of measured prompts was normalized to activity and correlated to body weight, mass per body length and body mass index to find the best predicting parameter. Next, a patient-specific activity was derived and subsequently validated in 119 additional patients. Image quality was scored by three experts on a four-point scale.Results. Both image quality and prompts decreased in heavier patients when using a fixed activity (p < .005). Body weight was used to derive a new activity formula: Activity = 8.3 MBq/ kg. When applying this formula, both measured prompts and scored image quality became independent of body weight (p > .60).Conclusion. Administrating a Rb-82 activity that linearly depends on body weight resulted in a constant image quality across all patients and is recommended. (J Nucl Cardiol 2019

No abstract